Research On Optimal Control Method For Fermentation Process Based On Cooperative Particle Swarm Optimization Algorithm

The optimal control of fermentation process is the effective way to improve the fermentation production quality and achieve economic benefits. Traditional optimization control method in solving complex problems is difficult to work out the ideal solution. With the development of artificial intelligence technology, multi-objective intelligent optimization algorithm which has global and multidirectional search process is beneficial to achieve the best balance of each index in the process of fermentation. Particle swarm optimization algorithm is suitable for solving complex problems because of simple principle, fast convergence speed, less set parameters thus become a research hotspot at home and abroad. Collaborative search method through information communication between swarms is helpful to improve the convergence speed and enhance the global search ability of particle swarm algorithm. Therefore, research on the optimal control method for fermentation process based on cooperative particle swarm optimization algorithm has important theoretical significance and application value.A context vector updating based particle swarm optimization algorithm is proposed through analysing the cooperatin mechanism of particle swarm optimization algorithm and the effectiveness is tested. Through analysing the multi-objective decomposition strategy and search method of particle, a cooperative multi-objective particle swarm optimization algorithm based on the decomposition is presented, and a kind of dynamic weight vector method is introduced which is to improve the diversity and distribution of the non inferior solution set. Based on this, context vector updating based particle swarm optimization algorithm is applied to parameter optimization in penicillin fermentation process model, and the industrial yeast fermentation process feed rate control have been optimized by using the cooperative multi-objective particle swarm optimization algorithm based on the decomposition.Experimental research shows that context vector updating based particle swarm optimization algorithm can get more convergent test function fitness values and has good stability. In comparison of the basic multi-objective particle swarm optimization algorithm based on decomposition, the cooperative multi-objective particle swarm optimization algorithm based on the decomposition has better convergence and distribution. The fitted curve which is close to original curve in the parameter optimization of penicillin fermentation process model is obtained by context vector updating based particle swarm optimization algorithm. The industrial yeast fermentation process optimization control method based on multi-objective particle swarm optimization algorithm get a reasonable feed rate curve, which provides an effective way for the optimal control of fermentation process.